This proposal describes a training plan for the candidate by an interdisciplinary group of experts in pancreatic cancer, cancer epidemiology, proteomics, and biostatistical analysis. The mentoring process will include epidemiologic and advanced statistical coursework, critical reading sessions with mentors and co-mentors, training in grant writing and management, as well as the experience of the hands on research. The research proposed is to develop early biomarkers of pancreatic cancer. Pancreatic cancer is an almost uniformly lethal disease because the cancer can not be diagnosed at an early, curable stage. Moreover, once the cancer has formed, chemotherapy offers only minimal improvement in survival. What is needed is a better understanding of how pancreatic cancer forms, and to exploit that knowledge to improve early diagnosis. The study proposes to use proteomics technology to identify the proteins that are specific to pancreatic cancer. In preliminary studies, 30 proteins were shown to have abundance changes of at least two-fold in pancreatic cancer juice compared to normal controls. Validation studies, to date, suggest that the proteomics measurements are robust and accurate. The proteins have been analyzed with regard to function; nearly one quarter of the proteins are unique and their function has not yet been determined. This study will provide the basis for serum biomarker development for early detection of pancreatic cancer in the general population-the goal of this K07 application. The study has the potential to significantly change the outcome in a disease against which we have made minimal headway in the past 50 years. In addition, by the end of the study period the candidate will have transitioned into a well-trained, independent investigator.